ANN-based algorithm for estimation and compensation of lateral misalignment in dynamic wireless power transfer systems for EV charging

In this paper, an Artificial Neural Network (ANN) is employed for the estimation of LaTeral Misalignment (LTM) as well as compensation of its effect on Dynamic Wireless Power Transfer (DWPT) systems for Electric Vehicles (EVs) charging. In a DWPT system, energy efficiency and energy transfer capability are significantly affected by the degree of LTM. Therefore, the real-time estimation of LTM, followed by certain remedial actions, could result in higher energy efficiency and energy transfer capability of the system. To this aim, an ANN-based algorithm is proposed to estimate the LTM value. The controller samples the DC-link current and based on the changes in the current profile and the vehicle speed, the ANN estimates the LTM value. Finally, the controller sets a modified reference value for the primary coil current which compensates the reduction in energy transfer capability caused by LTM. Using the suggested method, without any extra equipment in the DWPT system, the expected value of transferred energy is increased by 32%. The proposed ANN is trained for 9 different speeds in the range of 10 to 50 km/h and 11 different degrees of LTM in the range from 0 to 60%. The inputs to the proposed ANN are the DC-link current, the integral of DC-link current, and the vehicle speed.